Pancreas Segmentation Using U-Net Based Segmentation Networks in CT Modality: A Comparative Analysis

Year 2022, , 94 - 98, 30.09.2022

Alperen Derin , Caglar Gurkan , Abdulkadir Budak , Hakan Karataş

https://doi.org/10.31590/ejosat.1171803

Cited By: 2

Abstract

The pancreas is one of the small size organs in the abdomen. Moreover, anatomical differences make it difficult to detect the pancreas. This project aims to automatically segmentation of pancreas. For this purpose, NIH-CT82 data set, which includes CT images from 82 patients was used. U-Net which is state-of-the-art model and its different versions, namely Attention U-Net, Residual U-Net, Attention Residual U-Net, and Residual U-Net++ were tested. Best predict performance was achieved by Residual U-Net with the dice of 0.903, IoU of 0.823, sensitivity of 0.898, specificity of 1.000, precision of 0.908, and accuracy of 0.999. Consequently, an artificial intelligence (AI) supported decision support system was created for pancreas segmentation.

Keywords

Pancreas, Segmentation, Deep learning, U-Net, Residual U-Net++

Thanks

This paper has been prepared by AKGUN Computer Incorporated Company. We would like to thank AKGUN Computer Inc. for providing all kinds of opportunities and funds for the execution of this project.

CT Modalitesinde U-Net Tabanlı Segmentasyon Ağlarını Kullanarak Pankreas Segmentasyonu: Karşılaştırmalı Bir Analiz

Abstract

Pankreas, karın içindeki küçük boyutlu organlardan biridir. Ayrıca anatomik farklılıklar, pankreasın tespit edilmesini oldukça zorlaştırmaktadır. Bu proje pankreasın otomatik olarak segmentasyonunu amaçlamaktadır. Bu amaçla 82 hastanın bilgisayarlı tomografi (BT) görüntülerini içeren NIH-CT82 veri seti kullanılmıştır. Son teknoloji bir model olan U-Net ve farklı versiyonları olan Attention U-Net, Residual U-Net, Attention Residual U-Net ve Residual U-Net++ test edilmiştir. En iyi tahmin performansı, 0.903 zar skoru, 0.823 IoU, 0.898 duyarlılık, 1.000 özgülüllük, 0.908 kesinlik ve 0.999 doğruluk ile Residual U-Net tarafından elde edilmiştir. Sonuç olarak pankreas segmentasyonu için yapay zeka (YZ) destekli bir karar destek sistemi oluşturulmuştur.

Keywords

Pankreas, Segmentasyon, Derin öğrenme, U-Net, Residual U-Net++

References

• Hu, J. X., Lin, Y. Y., Zhao, C. F., Chen, W. B., Liu, Q. C., Li, Q. W., & Gao, F. (2021). Pancreatic cancer: A review of epidemiology, trend, and risk factors. World Journal of Gastroenterology, 27(27), 4298. https://doi.org/10.3748/WJG.V27.I27.4298
• Chaudhary, V., & Bano, S. (2011). Imaging of the pancreas: Recent advances. Indian Journal of Endocrinology and Metabolism, 15(5), 25. https://doi.org/10.4103/2230-8210.83060
• Liu, Z., Su, J., Wang, R., Jiang, R., Song, Y. Q., Zhang, D., Zhu, Y., Yuan, D., Gan, Q., & Sheng, V. S. (2022). Pancreas Co-segmentation based on dynamic ROI extraction and VGGU-Net. Expert Systems with Applications, 192, 116444. https://doi.org/10.1016/j.eswa.2021.116444
• Zhang, D., Zhang, J., Zhang, Q., Han, J., Zhang, S., & Han, J. (2021). Automatic pancreas segmentation based on lightweight DCNN modules and spatial prior propagation. Pattern Recognition, 114, 107762. https://doi.org/10.1016/j.patcog.2020.107762
• Dogan, R. O., Dogan, H., Bayrak, C., & Kayikcioglu, T. (2021). A Two-Phase Approach using Mask R-CNN and 3D U-Net for High-Accuracy Automatic Segmentation of Pancreas in CT Imaging. Computer Methods and Programs in Biomedicine, 207, 106141. https://doi.org/10.1016/j.cmpb.2021.106141
• Liu, Z., Su, J., Wang, R., Jiang, R., Song, Y. Q., Zhang, D., Zhu, Y., Yuan, D., Gan, Q., & Sheng, V. S. (2022). Pancreas Co-segmentation based on dynamic ROI extraction and VGGU-Net. Expert Systems with Applications, 192, 116444. https://doi.org/10.1016/J.ESWA.2021.116444
• Yan, Y., & Zhang, D. (2021). Multi-scale U-like network with attention mechanism for automatic pancreas segmentation. PLOS ONE, 16(5), e0252287. https://doi.org/10.1371/JOURNAL.PONE.0252287
• Li, M., Lian, F., Wang, C., & Guo, S. (2021). Accurate pancreas segmentation using multi-level pyramidal pooling residual U-Net with adversarial mechanism. BMC Medical Imaging, 21(1), 1–8. https://doi.org/10.1186/S12880-021-00694-1/FIGURES/5
• Cai, J., Lu, L., Xie, Y., Xing, F., & Yang, L. (2017). Improving Deep Pancreas Segmentation in CT and MRI Images via Recurrent Neural Contextual Learning and Direct Loss Function. https://doi.org/10.48550/arxiv.1707.04912
• Roth, H. R., Lu, L., Farag, A., Shin, H. C., Liu, J., Turkbey, E. B., & Summers, R. M. (2015). Deeporgan: Multi-level deep convolutional networks for automated pancreas segmentation. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 9349, 556–564. https://doi.org/10.1007/978-3-319-24553-9_68/COVER
• Ronneberger, O., Fischer, P., & Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 9351, 234–241. https://doi.org/10.1007/978-3-319-24574-4_28/COVER
• Oktay, O., Schlemper, J., Folgoc, L. Le, Lee, M., Heinrich, M., Misawa, K., Mori, K., McDonagh, S., Hammerla, N. Y., Kainz, B., Glocker, B., & Rueckert, D. (2018). Attention U-Net: Learning Where to Look for the Pancreas. https://doi.org/10.48550/arxiv.1804.03999
• Zhang, Z., Liu, Q., & Wang, Y. (2018). Road Extraction by Deep Residual U-Net. IEEE Geoscience and Remote Sensing Letters, 15(5), 749–753. https://doi.org/10.1109/LGRS.2018.2802944
• Chen, X., Yao, L., & Zhang, Y. (2020). Residual Attention U-Net for Automated Multi-Class Segmentation of COVID-19 Chest CT Images. https://doi.org/10.48550/arxiv.2004.05645
• Jha, D., Smedsrud, P. H., Riegler, M. A., Johansen, D., De Lange, T., Halvorsen, P., & Johansen, H. D. (2019). ResUNet++: An Advanced Architecture for Medical Image Segmentation. Proceedings - 2019 IEEE International Symposium on Multimedia, ISM 2019, 225–230. https://doi.org/10.1109/ISM46123.2019.00049
• Iqbal, H. (2018). Harisiqbal88/plotneuralnet v1. 0.0. URL: https://doi. org/10.5281/Zenodo.